skills.rest
maragudk
maragudkOFFICIAL
Explore all skills by this author →
WebsiteGitHub

unsloth

Fine-tune open LLMs on a single GPU with memory-efficient workflows.

27|2|Updated Mar 27, 2026
Search Tags:#embeddings#llm#fine-tuning#vision#gpu#tts#rl
One-click install
npx skills add https://github.com/maragudk/fabrik --skill unsloth-maragudk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: unsloth
Source: https://github.com/maragudk/fabrik/tree/main/skills/unsloth
Command: npx skills add https://github.com/maragudk/fabrik --skill unsloth-maragudk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fine-tune open LLMs and related models on a single GPU with memory-efficient workflows using Unsloth.

Core Features & Use Cases

  • End-to-end SFT with LoRA/QLoRA for decoder models.
  • RL fine-tuning (GRPO, DPO, ORPO, KTO, SimPO) with verifiable rewards.
  • Embedding fine-tuning using SentenceTransformer pipelines.
  • Vision and TTS fine-tuning and export options (GGUF, vLLM) for deployment.
  • Cross-framework compatibility with Hugging Face Transformers, TRL, PEFT.

Quick Start

Install Unsloth, choose a compatible model, and start a single-GPU fine-tuning run using the provided workflow.

Frequently Asked Questions about unsloth

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fine-tune an LLM on a single GPU without running out of memory?▼

You can fine-tune open LLMs on a single GPU using memory-efficient workflows with Unsloth. It provides SFT, RL, embeddings, and vision/TTS workflows to prevent out-of-memory errors.

What reinforcement learning methods are supported for LLM fine-tuning?▼

Supported RL fine-tuning methods include GRPO, DPO, ORPO, KTO, and SimPO. These workflows support verifiable rewards to align model outputs effectively.

Can I use Hugging Face Transformers, TRL, and PEFT for model training?▼

Yes, single-GPU fine-tuning supports cross-framework compatibility with Hugging Face Transformers, TRL, and PEFT. It provides end-to-end SFT with LoRA/QLoRA integrations loaded from the skill documentation.

How do I export fine-tuned models for deployment with vLLM or GGUF?▼

After fine-tuning, you can export models to deployment formats like GGUF and vLLM. This supports various workflows including vision and TTS model fine-tuning for production environments.

Does single-GPU fine-tuning work for embedding and vision models?▼

Yes, single-GPU fine-tuning supports SentenceTransformer pipelines for embedding fine-tuning alongside dedicated vision and TTS workflows. These utilize memory-efficient training across all model types.

Related Skills

Explore More Software Engineering Skills→
hhhi21ghhhi21g

unsloth

Fine-tune AI models with Unsloth for faster training and reduced memory usage.

Community
Advanced
ScientiaCapitalScientiaCapital
1

model-deployment

Export fine-tuned models to GGUF and deploy with vLLM or Ollama.

Community
Advanced
Marwan78888Marwan78888
16

fine-tuning-expert

Configure LoRA/QLoRA fine-tuning workflows with JSONL datasets and hyperparameters.

Community
Advanced
Tommy-ywTommy-yw
539

peft-fine-tuning

Fine-tune large language models with parameter-efficient methods like LoRA and QLoRA.

Community
Advanced